Red blood cell traits : drivers and consequences for cardiovascular disease and lifespan

Abstract

Cardiovascular disease (CVD) is the leading cause of global mortality. Currently, CVD is more common in men than women and so contributes to considerably shorter lives in men than women, despite ongoing improvements in CVD prevention and treatment. Historically, this disparity was explained in terms of estrogen protecting women, however, trials of estrogen in men and women many years ago did not substantiate that explanation. Less attention has been paid to the alternative possibility that substantially higher testosterone in men than women is putting men are at higher risk of CVD than women. However, use of testosterone by transmen has begun to suggest that transmen are at similar risk of CVD and mortality as cismen. There is little evidence from randomized controlled trials (RCT) in men that testosterone affects key CVD risk factors, such as blood pressure or lipids, while RCT evidence concerning the role of testosterone in ciswomen is lacking. To address this conundrum, I used Mendelian randomization applied to the largest available relevant genome wide association studies, I investigated the role of the most active form of testosterone (free testosterone in men and total testosterone in women) in emerging CVD risk factors potentially driven by testosterone, specifically reticulocytes and some fatty acids, and then of some of these factors in ischemic heart disease (IHD) from Cardiogram (cases=60,801, controls=123,504) and FinnGen (cases=31,640, controls=187,152) and lifespan (maternal n=412,937 and paternal n=415,311). Specifically, I found using the UK Biobank that testosterone in men (n=178,782, 118 genetic instruments) raised reticulocyte count 0.08 standard deviation (SD) per SD higher free testosterone, but total testosterone in women (n=230,454, 254 genetic instruments) did not -0.02 SD per SD higher total testosterone using inverse variance weighted estimates, with similar estimates from other methods. Consistent with a previous study I found that reticulocyte count was associated with higher risk of IHD (odds ratio 1.13 per SD reticulocytes, 95% CI 1.06 to 1.21 for Cardiogram and 1.13, 95% CI 1.05 to 1.22 for FinnGen), with similar estimates after allowing for red blood cell count and haemoglobin. I also found reticulocyte count associated with shorter life (-0.54 years, 95% CI -0.83 to -0.27), with similar estimates after allowing for red blood cell count and haemoglobin. Finally, in a two-step univariable and multivariable MR study, I showed that reticulocyte count, but not red blood cell count or haemoglobin, might partly mediated the effects of some fatty acids (polyunsaturated fatty acids (PUFA), specifically omega-6 and linoleic acid) on IHD and lifespan. Given, an MR study has recently shown that PUFA may increase risk of IHD and shorten lifespan particularly in men, these findings start to provide an explanation for higher rates of IHD and shorter lives in men than women. Whether reticulocytes are a possible target for interventions to mitigate IHD and increase lifespan particularly in men, or whether they are a downstream consequence of an upstream, possibly sex-specific driver, of CVD and lifespan should be considered, so as to identify additional intervention targets to address disparities in IHD and lifespan.